Carlo Rubbia: Muon Cooling at CERN
Rubbia started his talk on a muon collider option for Higgs physics claiming that we need precision to search for deviations from SM in the Higgs sector. He recalled how indirect predictions on the top and Higgs mass were obtained from electroweak fits. A similar situation can lead in the future to new discoveries if we measure with high precision the Higgs boson properties.
The scalar sector is one of the keys to our future understanding of HEP. So a second phase of studies after the discovery is necessary for the Higgs boson. There are two alternatives:
– an e+e- collider with L>10^34 to measure ZH (xs=200fb), with a circumference
of 80km, or 31km length of a linear collider.
– a mu+mu- collider at L>10^32 (xs=20 pb), radius of only 50m, but novel muon cooling technology is needed.
Many bids for a circular e+e- have been put forth in the US, Europe, and Asia. For instance a Super Tristan in Japan, with a 80km circumference. Also CERN management is supporting the 80km tunnel to be built around Geneva.
There are problems to build such a huge machine. The luminosity has to be pushed to the beam-strahlung limit; collisions occur at an angle, with fewer bunches than for a B factory. Power costs are comparable to those of a linear collider (100 MW). The extremely small vertical emittance required, with a beam crossing size of 0.01 mucrons, i.e. 300 times smaller than that of LEP 2. Rubbia claims that the performance is at the border of feasibility. In any case the H width of 4.5 MeV cannot be directly observed.
The ILC option has also been discussed by a large group of physicists. Cost is comparable to that of a circular machine, as is power consumption. The more conservative ring option appears preferable.
Rubbia showed a plot of coupling measurements predicted for LHC vs ILC at various energies, making the point that to make a 5-sigma discovery of anomalous couplings one should aim to a 1% precision, which is not granted by a linear collider.
The s-channel H production in muon collisions is enhanced by the mass squaed ratio. This is the cleanest production mode. S-channel production has many advantages, to distinguish a SM from a SM-like H from SUSY.
Rubbia then went into the technical details of the proposed machine. He showed that the machine could be easily housed within CERN, with an accelerator structure including two additional small storage ring with R=50m, stripping H ions to a tight proton bunch, compressing it to few nanoseconds. The muons must then be cooled and focused in a B=20T field. Muons collide in a storage ring of 60m radius.
As muons are created from the decay in flight of secondary pions downstream the target, they have a large spread in energy and a small spread in time. Two techniques can be used to achieve a small energy spread with larger time spread, achieving a manageable bunched beam. Simulations show that a nice bunching can be achieved with 10% spread in momentum, and muons of both signs can be collected. One may collect more than ten trillion muons in the best 12 bunches, using 50 pulses per second and 100 trillion protons per pulse.
The heart of the discussion is to understand how to cool the muon beam. It can be done with dE/dx coling, closely resembling the synchrotron compression of relativistic electrons. Muons undergo multiple energy losses in thin, low-Z absorber. This produces a very fast cooling, which is essential for the muons (the muon lifetime is 2.2 microseconds).
The proposed scheme (D.V.Neuffer, NIM A 350 (1994) 27) with a wedge of absorber is claimed to produce the wanted effect for muon momenta of 250 MeV. Rubbia showed a complex graph (fig.1, above) of transverse versus longitudinal emittance, where initially there is a purely transverse cooling, performed for both muon signs by a linear buncher and rotator; then there is a ring which does most of the work, cooling both transversely and longitudinally. Then the bunching increases the longitudinal emittance, a further cooling then brings to the final configuration, the PIC resonance cooling, which further reduces the transverse emittance by over an order of magnitude with no change in longitudinal emittance. This last step is the critical part. Finally one should achieve 0.04 pi mmrad transverse emittance and 1 pi mm rad longitudinal emittance.
The PIC (see fig.2 above) is combining ionization colling with parametric resonances. A half-integer resonance is induced in the beam, such that the normal elliptical motion of particles in x-x’ phase space becomes hyperbolic, with particles moving to smaller x and larger x’ at the channel focal points. It is totally unstable in the absence of cooling, but with absorbers you can find an equilibrium (V.S.Morozov et al., AIP 1507, 843 (2012)). Longitudinal emittance is kept constant tapering the absorbers and placing them at points of appropriate dispersion.
All in all, the estimated performance of the H factory is a luminosity of 5×10^32 cm-2 s-1, with estimated 44,000 events per year. This is done by using a trillion muons per bunch, at a proton power of 4 MW (an energy of 5 GeV per proton).
Rubbia wants to make soon the next step, which should be the one of experimentally demonstrating the 6D cooling and the PIC cooling which are critical for the muon collider project. Most of the important work is to develop a resonant cooling. THis may involve significant and unexpected conditions which are hard to predict without a practical test.
In conclusion, Rubbia stressed that in his view the recent discovery of the Higgs puts us in the situation of having to investigate its properties in detail. A high-energy muon collider is the only possible circular high-energy lepton Higgs collider that can be easily situated within the existing CERN or FNAL sites.